Provided by: libtest-deep-perl_1.128-1_all bug


       Test::Deep - Extremely flexible deep comparison


         use Test::More tests => $Num_Tests;
         use Test::Deep;

           "got the right horrible nested data structure"

           methods(name => "John", phone => "55378008"),
           "object methods ok"

           [$hash1, $hash2, ignore()],
           "first 2 elements are as expected, ignoring 3"

           noclass({value => 5}),
           "object looks ok, not checking its class"

           bag('a', 'b', {key => [1, 2]}),
           "array has the 3 things we wanted in some order"


       If you don't know anything about automated testing in Perl then you should probably read
       about Test::Simple and Test::More before preceding.  Test::Deep uses the Test::Builder

       Test::Deep gives you very flexible ways to check that the result you got is the result you
       were expecting. At its simplest it compares two structures by going through each level,
       ensuring that the values match, that arrays and hashes have the same elements and that
       references are blessed into the correct class. It also handles circular data structures
       without getting caught in an infinite loop.

       Where it becomes more interesting is in allowing you to do something besides simple exact
       comparisons. With strings, the "eq" operator checks that 2 strings are exactly equal but
       sometimes that's not what you want. When you don't know exactly what the string should be
       but you do know some things about how it should look, "eq" is no good and you must use
       pattern matching instead. Test::Deep provides pattern matching for complex data structures

       Test::Deep has a lot of exports.  See "EXPORTS" below.


       How Test::Deep works is much easier to understand by seeing some examples.

   Without Test::Deep
       Say you want to test a function which returns a string. You know that your string should
       be a 7 digit number beginning with 0, "eq" is no good in this situation, you need a
       regular expression. So you could use Test::More's "like()" function:

         like($string, qr/^0[0-9]{6}$/, "number looks good");

       Similarly, to check that a string looks like a name, you could do:

         like($string, qr/^(Mr|Mrs|Miss) \w+ \w+$/,
           "got title, first and last name");

       Now imagine your function produces a hash with some personal details in it.  You want to
       make sure that there are 2 keys, Name and Phone and that the name looks like a name and
       the phone number looks like a phone number. You could do:

         $hash = make_person();
         like($hash->{Name}, qr/^(Mr|Mrs|Miss) \w+ \w+$/, "name ok");
         like($hash->{Phone}, qr/^0[0-9]{6}$/, "phone ok");
         is(scalar keys %$hash, 2, "correct number of keys");

       But that's not quite right, what if make_person has a serious problem and didn't even
       return a hash? We really need to write

         if (ref($hash) eq "HASH")
           like($hash->{Name}, qr/^(Mr|Mrs|Miss) \w+ \w+$/, "name ok");
           like($hash->{Phone}, qr/^0[0-9]{6}$/, "phone ok");
           is(scalar keys %$hash, 2, "correct number of keys");
           fail("person not a hash");
           fail("person not a hash");
           fail("person not a hash"); # need 3 to keep the plan correct

       Already this is getting messy, now imagine another entry in the hash, an array of
       children's names. This would require

         if (ref($hash) eq "HASH")
           like($hash->{Name}, $name_pat, "name ok");
           like($hash->{Phone}, '/^0d{6}$/', "phone ok");
           my $cn = $hash->{ChildNames};
           if (ref($cn) eq "ARRAY")
             foreach my $child (@$cn)
               like($child, $name_pat);
               fail("child names not an array")
           fail("person not a hash");

       This is a horrible mess and because we don't know in advance how many children's names
       there will be, we can't make a plan for our test anymore (actually, we could but it would
       make things even more complicated).

       Test::Deep to the rescue.

   With Test::Deep
         my $name_re = re('^(Mr|Mrs|Miss) \w+ \w+$');
             Name => $name_re,
             Phone => re('^0d{6}$'),
             ChildNames => array_each($name_re)
           "person ok"

       This will do everything that the messy code above does and it will give a sensible message
       telling you exactly what went wrong if it finds a part of $person that doesn't match the
       pattern. "re()" and "array_each()" are special function imported from Test::Deep. They
       create a marker that tells Test::Deep that something different is happening here. Instead
       of just doing a simple comparison and checking are two things exactly equal, it should do
       something else.

       If a person was asked to check that 2 structures are equal, they could print them both out
       and compare them line by line. The markers above are similar to writing a note in red pen
       on one of the printouts telling the person that for this piece of the structure, they
       should stop doing simple line by line comparison and do something else.

       "re($regex)" means that Test::Deep should check that the current piece of data matches the
       regex in $regex. "array_each($struct)" means that Test::Deep should expect the current
       piece of data to be an array and it should check that every element of that array matches
       $struct.  In this case, every element of "$person->{ChildNames}" should look like a name.
       If say the 3rd one didn't you would get an error message something like

         Using Regexp on $data->{ChildNames}[3]
            got    : 'Queen John Paul Sartre'
            expect : /^(Mr|Mrs|Miss) \w+ \w+$/

       There are lots of other special comparisons available, see "SPECIAL COMPARISONS PROVIDED"
       below for the full list.

   Reusing structures
       Test::Deep is good for reusing test structures so you can do this

         my $name_re = re('^(Mr|Mrs|Miss) \w+ \w+$');
         my $person_cmp = {
           Name => $name_re,
           Phone => re('^0d{6}$'),
           ChildNames => array_each($name_re)

         cmp_deeply($person1, $person_cmp, "person ok");
         cmp_deeply($person2, $person_cmp, "person ok");
         cmp_deeply($person3, $person_cmp, "person ok");

       You can even put $person_cmp in a module and let other people use it when they are writing
       test scripts for modules that use your modules.

       To make things a little more difficult, lets change the person data structure so that
       instead of a list of ChildNames, it contains a list of hashes, one for each child. So in
       fact our person structure will contain other person structures which may contain other
       person structures and so on.  This is easy to handle with Test::Deep because Test::Deep
       structures can include themselves. Simply do

         my $name_re = re('^(Mr|Mrs|Miss) \w+ \w+$');
         my $person_cmp = {
           Name => $name_re,
           Phone => re('^0d{6}$'),
           # note no mention of Children here

         $person_cmp->{Children} = array_each($person_cmp);

         cmp_deeply($person, $person_cmp, "person ok");

       This will now check that $person->{Children} is an array and that every element of that
       array also matches $person_cmp, this includes checking that its children also match the
       same pattern and so on.

   Circular data structures
       A circular data structure is one which loops back on itself, you can make one easily by

         my @b;
         my @a = (1, 2, 3, \@b);
         push(@b, \@a);

       now @a contains a reference to be @b and @b contains a reference to @a. This causes
       problems if you have a program that wants to look inside @a and keep looking deeper and
       deeper at every level, it could get caught in an infinite loop looking into @a then @b
       then @a then @b and so on.

       Test::Deep avoids this problem so we can extend our example further by saying that a
       person should also list their parents.

         my $name_re = re('^(Mr|Mrs|Miss) \w+ \w+$');
         my $person_cmp = {
           Name => $name_re,
           Phone => re('^0d{6}$'),
           # note no mention of Children here

         $person_cmp->{Children} = each_array($person_cmp);
         $person_cmp->{Parents} = each_array($person_cmp);

         cmp_deeply($person, $person_cmp, "person ok");

       So this will check that for each child $child in "$person->{Children}" that the
       "$child->{Parents}" matches $person_cmp however it is smart enough not to get caught in an
       infinite loop where it keeps bouncing between the same Parent and Child.


       "cmp_deeply($got, $expected, $name)" takes 3 arguments. $got is the structure that you are
       checking, you must not include any special comparisons in this structure or you will get a
       fatal error. $expected describes what Test::Deep will be looking for in $got. You can put
       special comparisons in $expected if you want to.

       As Test::Deep descends through the 2 structures, it compares them one piece at a time, so
       at any point in the process, Test::Deep is thinking about 2 things - the current value
       from $got and the current value from $expected. In the documentation, I call them $got_v
       and "exp_v" respectively.



         my $ok = cmp_deeply($got, $expected, $name)

       $got is the result to be checked. $expected is the structure against which $got will be
       check. $name is the test name.

       This is the main comparison function, the others are just wrappers around this.  $got and
       $expected are compared recursively.  Each value in $expected defines what's expected at
       the corresponding location in $got.  Simple scalars are compared with "eq".  References to
       structures like hashes and arrays are compared recursively.

       Items in $expected, though, can also represent complex tests that check for numbers in a
       given range, hashes with at least a certain set of keys, a string matching a regex, or
       many other things.

       See "WHAT ARE SPECIAL COMPARISONS" for details.


         my $ok = cmp_bag(\@got, \@bag, $name)

       Is shorthand for cmp_deeply(\@got, bag(@bag), $name)

       n.b.: Both arguments must be array refs. If they aren't an exception will be thrown.


         my $ok = cmp_set(\@got, \@set, $name)

       Is shorthand for cmp_deeply(\@got, set(@set), $name)


         my $ok = cmp_methods(\@got, \@methods, $name)

       Is shorthand for cmp_deeply(\@got, methods(@methods), $name)


         my $ok = eq_deeply($got, $expected)

       This is the same as cmp_deeply() except it just returns true or false. It does not create
       diagnostics or talk to Test::Builder, but if you want to use it in a non-testing
       environment then you should import it through Test::Deep::NoTest. For example

         use Test::Deep::NoTest;
         print "a equals b" unless eq_deeply($a, $b);

       otherwise the Test::Builder framework will be loaded and testing messages will be output
       when your program ends.


         ($ok, $stack) = cmp_details($got, $expected)

       This behaves much like eq_deeply, but it additionally allows you to produce diagnostics in
       case of failure by passing the value in $stack to "deep_diag".

       Do not make assumptions about the structure or content of $stack and do not use it if $ok
       contains a true value.

       See "USING TEST::DEEP WITH TEST::BUILDER" for example uses.


       In the documentation below, $got_v is used to indicate any given value within the $got


         cmp_deeply( $got, ignore() );

       This makes Test::Deep skip tests on $got_v. No matter what value $got_v has, Test::Deep
       will think it's correct. This is useful if some part of the structure you are testing is
       very complicated and already tested elsewhere, or if it is unpredictable.

             name    => 'John',
             random  => ignore(),
             address => [ '5 A street', 'a town', 'a country' ],

       is the equivalent of checking

         $got->{name} eq 'John';
         exists $got->{random};
         cmp_deeply($got->{address}, ['5 A street', 'a town', 'a country']);


         cmp_deeply( $got, methods(%hash) );

       %hash is a hash of method call => expected value pairs.

       This lets you call methods on an object and check the result of each call.  The methods
       will be called in the order supplied. If you want to pass arguments to the method you
       should wrap the method name and arguments in an array reference.

           methods(name => "John", ["favourite", "food"] => "taco")

       is roughly the equivalent of checking that

         $obj->name eq "John"
         $obj->favourite("food") eq "taco"

       The methods will be called in the order you supply them and will be called in scalar
       context. If you need to test methods called in list context then you should use

       NOTE Just as in a normal test script, you need to be careful if the methods you call have
       side effects like changing the object or other objects in the structure. Although the
       order of the methods is fixed, the order of some other tests is not so if $expected is

           manager => methods(@manager_methods),
           coder => methods(@coder_methods)

       there is no way to know which if manager and coder will be tested first. If the methods
       you are testing depend on and alter global variables or if manager and coder are the same
       object then you may run into problems.


         cmp_deeply( $got, listmethods(%hash) );

       %hash is a hash of pairs mapping method names to expected return values.

       This is almost identical to methods() except the methods are called in list context
       instead of scalar context. This means that the expected return values supplied must be in
       array references.

             name => "John",
             ["favourites", "food"] => ["Mapo tofu", "Gongbao chicken"]

       is the equivalent of checking that

         $obj->name eq "John"
         cmp_deeply([$obj->favourites("food")], ["Mapo tofu", "Gongbao chicken"]);

       The methods will be called in the order you supply them.

       NOTE The same caveats apply as for methods().


         cmp_deeply( $got, shallow($thing) );

       $thing is a ref.

       This prevents Test::Deep from looking inside $thing. It allows you to check that $got_v
       and $thing are references to the same variable. So

         my @a = @b = (1, 2, 3);
         cmp_deeply(\@a, \@b);

       will pass because @a and @b have the same elements however

         cmp_deeply(\@a, shallow(\@b))

       will fail because although "\@a" and "\@b" both contain "1, 2, 3" they are references to
       different arrays.


         cmp_deeply( $got, noclass($thing) );

       $thing is a structure to be compared against.

       This makes Test::Deep ignore the class of objects, so it just looks at the data they
       contain. Class checking will be turned off until Test::Deep is finished comparing $got_v
       against $thing. Once Test::Deep comes out of $thing it will go back to its previous
       setting for checking class.

       This can be useful when you want to check that objects have been constructed correctly but
       you don't want to write lots of "bless"es. If @people is an array of Person objects then

         cmp_deeply(\@people, [
           bless {name => 'John', phone => '555-5555'}, "Person",
           bless {name => 'Anne', phone => '444-4444'}, "Person",

       can be replaced with

         cmp_deeply(\@people, noclass([
           {name => 'John', phone => '555-5555'},
           {name => 'Anne', phone => '444-4444'}

       However, this is testing so you should also check that the objects are blessed correctly.
       You could use a map to bless all those hashes or you could do a second test like

         cmp_deeply(\@people, array_each(isa("Person"));


         cmp_deeply( $got, useclass($thing) );

       This turns back on the class comparison while inside a "noclass()".

               useclass( $object )

       In this example the class of the array reference in $got is ignored but the class of
       $object is checked, as is the class of everything inside $object.


         cmp_deeply( $got, re($regexp, $capture_data, $flags) );

       $regexp is either a regular expression reference produced with "qr/.../" or a string which
       will be used to construct a regular expression.

       $capture_data is optional and is used to check the strings captured by an regex. This
       should can be an array ref or a Test::Deep comparator that works on array refs.

       $flags is an optional string which controls whether the regex runs as a global match. If
       $flags is "g" then the regex will run as "m/$regexp/g".

       Without $capture_data, this simply compares $got_v with the regular expression provided.

         cmp_deeply($got, [ re("ferg") ])

       is the equivalent of

         $got->[0] =~ /ferg/

       With $capture_data,

         cmp_deeply($got, [re($regex, $capture_data)])

       is the equivalent of

         my @data = $got->[0] =~ /$regex/;
         cmp_deeply(\@data, $capture_data);

       So you can do something simple like

         cmp_deeply($got, re(qr/(\d\d)(\w\w)/, [25, "ab" ]))

       to check that "(\d\d)" was 25 and "(\w\w)" was "ab" but you can also use Test::Deep
       objects to do more complex testing of the captured values

             set(qw( cat sheep dog )),

       here, the regex will match the string and will capture the animal names and check that
       they match the specified set, in this case it will fail, complaining that "goat" is not in
       the set.


         cmp_deeply( $got, all(@expecteds) );

       @expecteds is an array of expected structures.

       This allows you to compare data against multiple expected results and make sure each of
       them matches.

         cmp_deeply($got, all(isa("Person"), methods(name => 'John')))

       is equivalent to

         $got->name eq 'John'

       If either test fails then the whole thing is considered a fail. This is a short-circuit
       test, the testing is stopped after the first failure, although in the future it may
       complete all tests so that diagnostics can be output for all failures. When reporting
       failure, the parts are counted from 1.

       Thanks to the magic of overloading, you can write

         any( re("^wi"), all(isa("Person"), methods(name => 'John')) )


          re("^wi") | isa("Person") & methods(name => 'John')

       Note single "|" not double, as "||" cannot be overloaded. This will only work when there
       is a special comparison involved. If you write

         "john" | "anne" | "robert"

       Perl will turn this into


       which is presumably not what you wanted. This is because perl ors them together as strings
       before Test::Deep gets a chance to do any overload tricks.


         cmp_deeply( $got, any(@expecteds) );

       @expecteds is an array of expected structures.

       This can be used to compare data against multiple expected results and make sure that at
       least one of them matches. This is a short-circuit test so if a test passes then none of
       the tests after that will be attempted.

       You can also use overloading with "|" similarly to all().


         cmp_deeply( $got, Isa($class) );


         cmp_deeply( $got, isa($class) );

       $class is a class name.

       This uses "UNIVERSAL::isa()" to check that $got_v is blessed into the class $class.

       NOTE: "Isa()" does exactly as documented here, but "isa()" is slightly different. If
       "isa()" is called with 1 argument it falls through to "Isa()". If "isa()" called with 2
       arguments, it falls through to "UNIVERSAL::isa". This is to prevent breakage when you
       import "isa()" into a package that is used as a class. Without this, anyone calling
       "Class->isa($other_class)" would get the wrong answer. This is a hack to patch over the
       fact that "isa" is exported by default.


         cmp_deeply( $got, obj_isa($class) );

       This test accepts only objects that are instances of $class or a subclass.  Unlike the
       "Isa" test, this test will never accept class names.


         cmp_deeply( \@got, array_each($thing) );

       $thing is a structure to be compared against.

       <$got_v> must be an array reference. Each element of it will be compared to $thing. This
       is useful when you have an array of similar things, for example objects of a known type
       and you don't want to have to repeat the same test for each one.

         my $common_tests = all(
              handle => isa("IO::Handle")
              filename => re("^/home/ted/tmp"),

         cmp_deeply($got, array_each($common_tests));

       is similar to

         foreach my $got_v (@$got) {
           cmp_deeply($got_v, $common_tests)

       Except it will not explode if $got is not an array reference. It will check that each of
       the objects in @$got is a MyFile and that each one gives the correct results for its

       You could go further, if for example there were 3 files and you knew the size of each one
       you could do this

               methods(size => 1000),
               methods(size => 200),
               methods(size => 20)
         cmp_deeply($got, array_each($structure));


         cmp_deeply( \%got, hash_each($thing) );

       This test behaves like "array_each" (see above) but tests that each hash value passes its


         cmp_deeply( $got, str($string) );

       $string is a string.

       This will stringify $got_v and compare it to $string using "eq", even if $got_v is a ref.
       It is useful for checking the stringified value of an overloaded reference.


         cmp_deeply( $got, num($number, $tolerance) );

       $number is a number.

       $tolerance is an optional number.

       This will add 0 to $got_v and check if it's numerically equal to $number, even if $got_v
       is a ref. It is useful for checking the numerical value of an overloaded reference. If
       $tolerance is supplied then this will check that $got_v and $exp_v are less than
       $tolerance apart. This is useful when comparing floating point numbers as rounding errors
       can make it hard or impossible for $got_v to be exactly equal to $exp_v. When $tolerance
       is supplied, the test passes if "abs($got_v - $exp_v) <= $tolerance".

       Note in Perl, ""12blah" == 12" because Perl will be smart and convert "12blah" into 12.
       You may not want this. There was a strict mode but that is now gone. A "looks like a
       number" test will replace it soon. Until then you can usually just use the string()
       comparison to be more strict. This will work fine for almost all situations, however it
       will not work when <$got_v> is an overloaded value who's string and numerical values


         cmp_deeply( $got, bool($value) );

       $value is anything you like but it's probably best to use 0 or 1

       This will check that $got_v and $value have the same truth value, that is they will give
       the same result when used in boolean context, like in an "if()" statement.


         cmp_deeply( $got, code(\&subref) );

       "\&subref" is a reference to a subroutine which will be passed a single argument, it then
       should return a true or false and possibly a string

       This will pass $got_v to the subroutine which returns true or false to indicate a pass or
       fail. Fails can be accompanied by a diagnostic string which gives an explanation of why
       it's a fail.

         sub check_name
           my $name = shift;
           if ($boss->likes($name))
             return 1;
             return (0, "the boss doesn't like your name");

         cmp_deeply("Brian", code(\&check_name));

       Set comparisons give special semantics to array comparisons:

       ·   The order of items in a set is irrelevant

       ·   The presence of duplicate items in a set is ignored.

       As such, in any set comparison, the following arrays are equal:

         [ 1, 2 ]
         [ 1, 1, 2 ]
         [ 1, 2, 1 ]
         [ 2, 1, 1 ]
         [ 1, 1, 2 ]

       All are interpreted by "set" semantics as if the set was only specified as:

         [ 1, 2 ]

       All "set" functions return an object which can have additional items added to it:

         my $set = set( 1, 2 );
         $set->add(1, 3, 1 );  # Set is now ( 1, 2, 3 )

       Special care must be taken when using special comparisons within sets. See "SPECIAL CARE


         cmp_deeply( \@got, set(@elements) );

       This does a set comparison, that is, it compares two arrays but ignores the order of the
       elements and it ignores duplicate elements, but ensures that all items in in @elements
       will be in $got and all items in $got will be in @elements.

       So the following tests will be passes, and will be equivalent:

         cmp_deeply([1, 2, 2, 3], set(3, 2, 1, 1));
         cmp_deeply([1, 2, 3],    set(3, 2, 1));


         cmp_deeply( \@got, supersetof(@elements) );

       This function works much like "set", and performs a set comparison of $got_v with the
       elements of @elements.

       "supersetof" is however slightly relaxed, such that $got may contain things not in
       @elements, but must at least contain all @elements.

       These two statements are equivalent, and will be passes:

         cmp_deeply([1,2,3,3,4,5], supersetof(2,2,3));
         cmp_deeply([1,2,3,4,5],   supersetof(2,3));

       But these will be failures:

         cmp_deeply([1,2,3,4,5],   supersetof(2,3,6)); # 6 not in superset
         cmp_deeply([1],           supersetof(1,2));   # 2 not in superset


         cmp_deeply( \@got, subsetof(@elements) );

       This function works much like "set", and performs a set comparison of $got_v with the
       elements of @elements.

       This is the inverse of "supersetof", which expects all unique elements found in $got_v
       must be in @elements.

         cmp_deeply([1,2,4,5], subsetof(2,3,3)    ) # Fail: 1,4 & 5 extra
         cmp_deeply([2,3,3],   subsetof(1,2,4,5)  ) # Fail: 3 extra
         cmp_deeply([2,3,3],   subsetof(1,2,4,5,3)) # Pass


         cmp_deeply( $got, none(@elements) );

       @elements is an array of elements, wherein no elements in @elements may be equal to


         cmp_deeply( \@got, noneof(@elements) );

       @elements is an array of elements, wherein no elements in @elements may be found in

       For example:

         # Got has no 1, no 2, and no 3
         cmp_deeply( [1], noneof( 1, 2, 3 ) ); # fail
         cmp_deeply( [5], noneof( 1, 2, 3 ) ); # pass

       Bag comparisons give special semantics to array comparisons, that are similar to set
       comparisons, but slightly different.

       ·   The order of items in a bag is irrelevant

       ·   The presence of duplicate items in a bag is PRESERVED

       As such, in any bag comparison, the following arrays are equal:

         [ 1, 1, 2 ]
         [ 1, 2, 1 ]
         [ 2, 1, 1 ]
         [ 1, 1, 2 ]

       However, they are NOT equal to any of the following:

         [ 1, 2 ]
         [ 1, 2, 2 ]
         [ 1, 1, 1, 2 ]

       All "bag" functions return an object which can have additional items added to it:

         my $bag = bag( 1, 2 );
         $bag->add(1, 3, 1 );  # Bag is now ( 1, 1, 1, 2, 3 )

       Special care must be taken when using special comparisons within bags. See "SPECIAL CARE


         cmp_deeply( \@got, bag(@elements) );

       This does an order-insensitive bag comparison between $got and @elements, ensuring that:

       each item in @elements is found in $got
       the number of times a $expected_v is found in @elements is reflected in $got
       no items are found in $got other than those in @elements.

       As such, the following are passes, and are equivalent to each other:

         cmp_deeply([1, 2, 2], bag(2, 2, 1))
         cmp_deeply([2, 1, 2], bag(2, 2, 1))
         cmp_deeply([2, 2, 1], bag(2, 2, 1))

       But the following are failures:

         cmp_deeply([1, 2, 2],     bag(2, 2, 1, 1)) # Not enough 1's in Got
         cmp_deeply([1, 2, 2, 1],  bag(2, 2, 1)   ) # Too many   1's in Got


         cmp_deeply( \@got, superbagof( @elements ) );

       This function works much like "bag", and performs a bag comparison of $got_v with the
       elements of @elements.

       "superbagof" is however slightly relaxed, such that $got may contain things not in
       @elements, but must at least contain all @elements.


         # pass
         cmp_deeply( [1, 1, 2], superbagof( 1 )      );

         # fail: not enough 1's in superbag
         cmp_deeply( [1, 1, 2], superbagof( 1, 1, 1 ));


         cmp_deeply( \@got, subbagof(@elements) );

       This function works much like "bag", and performs a bag comparison of $got_v with the
       elements of @elements.

       This is the inverse of "superbagof", and expects all elements in $got to be in @elements,
       while allowing items to exist in @elements that are not in $got

         # pass
         cmp_deeply( [1],        subbagof( 1, 1, 2 ) );

         # fail: too many 1's in subbag
         cmp_deeply( [1, 1, 1],  subbagof( 1, 1, 2 ) );

       Typically, if you're doing simple hash comparisons,

         cmp_deeply( \%got, \%expected )

       is sufficient. "cmp_deeply" will ensure %got and %hash have identical keys, and each key
       from either has the same corresponding value.


         cmp_deeply( \%got, superhashof(\%hash) );

       This will check that the hash %$got is a "super-hash" of %hash. That is that all the key
       and value pairs in %hash appear in %$got but %$got can have extra ones also.

       For example

         cmp_deeply({a => 1, b => 2}, superhashof({a => 1}))

       will pass but

         cmp_deeply({a => 1, b => 2}, superhashof({a => 1, c => 3}))

       will fail.


         cmp_deeply( \%got, subhashof(\%hash) );

       This will check that the hash %$got is a "sub-hash" of %hash. That is that all the key and
       value pairs in %$got also appear in %hash.

       For example

         cmp_deeply({a => 1}, subhashof({a => 1, b => 2}))

       will pass but

         cmp_deeply({a => 1, c => 3}, subhashof({a => 1, b => 2}))

       will fail.



         my $reason = deep_diag($stack);

       $stack is a value returned by cmp_details.  Do not call this function if cmp_details
       returned a true value for $ok.

       "deep_diag()" returns a human readable string describing how the comparison failed.


       You've written a module to handle people and their film interests. Say you have a function
       that returns an array of people from a query, each person is a hash with 2 keys: Name and
       Age and the array is sorted by Name. You can do

             {Name => 'Anne', Age => 26},
             {Name => "Bill", Age => 47}
             {Name => 'John', Age => 25},

       Soon after, your query function changes and all the results now have an ID field. Now your
       test is failing again because you left out ID from each of the hashes. The problem is that
       the IDs are generated by the database and you have no way of knowing what each person's ID
       is. With Test::Deep you can change your query to

             {Name => 'John', Age => 25, ID => ignore()},
             {Name => 'Anne', Age => 26, ID => ignore()},
             {Name => "Bill", Age => 47, ID => ignore()}

       But your test still fails. Now, because you're using a database, you no longer know what
       order the people will appear in. You could add a sort into the database query but that
       could slow down your application. Instead you can get Test::Deep to ignore the order of
       the array by doing a bag comparison instead.

             {Name => 'John', Age => 25, ID => ignore()},
             {Name => 'Anne', Age => 26, ID => ignore()},
             {Name => "Bill", Age => 47, ID => ignore()}

       Finally person gets even more complicated and includes a new field called Movies, this is
       a list of movies that the person has seen recently, again these movies could also come
       back in any order so we need a bag inside our other bag comparison, giving us something

             {Name => 'John', Age => 25, ID => ignore(), Movies => bag(...)},
             {Name => 'Anne', Age => 26, ID => ignore(), Movies => bag(...)},
             {Name => "Bill", Age => 47, ID => ignore(), Movies => bag(...)}


       Combining "cmp_details" and "deep_diag" makes it possible to use Test::Deep in your own
       test classes.

       In a Test::Builder subclass, create a test method in the following form:

         sub behaves_ok {
           my $self = shift;
           my $expected = shift;
           my $test_name = shift;

           my $got = do_the_important_work_here();

           my ($ok, $stack) = cmp_details($got, $expected);
           unless ($Test->ok($ok, $test_name)) {
             my $diag = deep_diag($stack);

       As the subclass defines a test class, not tests themselves, make sure it uses
       Test::Deep::NoTest, not "Test::Deep" itself.


       Currently any CODE, GLOB or IO refs will be compared using shallow(), which means only
       their memory addresses are compared.


       There is a bug in set and bag compare to do with competing SCs. It only occurs when you
       put certain special comparisons inside bag or set comparisons you don't need to worry
       about it. The full details are in the "bag()" docs. It will be fixed in an upcoming


       If you use certain special comparisons within a bag or set comparison there is a danger
       that a test will fail when it should have passed. It can only happen if two or more
       special comparisons in the bag are competing to match elements.  Consider this comparison

         cmp_deeply(['furry', 'furball'], bag(re("^fur"), re("furb")))

       There are two things that could happen, hopefully "re("^fur")" is paired with "furry" and
       "re("^furb")" is paired with "furb" and everything is fine but it could happen that
       "re("^fur")" is paired with "furball" and then "re("^furb")" cannot find a match and so
       the test fails. Examples of other competing comparisons are "bag(1, 2, 2)" vs "set(1, 2)"
       and "methods(m1 => "v1", m2 => "v2")" vs "methods(m1 => "v1")"

       This problem is could be solved by using a slower and more complicated algorithm for set
       and bag matching. Something for the future...


       A special comparison (SC) is simply an object that inherits from Test::Deep::Cmp. Whenever
       $expected_v is an SC then instead of checking "$got_v eq $expected_v", we pass control
       over to the SC and let it do its thing.

       Test::Deep exports lots of SC constructors, to make it easy for you to use them in your
       test scripts. For example is "re("hello")" is just a handy way of creating a
       Test::Deep::Regexp object that will match any string containing "hello". So

         cmp_deeply([ 'a', 'b', 'hello world'], ['a', 'b', re("^hello")]);

       will check 'a' eq 'a', 'b' eq 'b' but when it comes to comparing 'hello world' and
       "re("^hello")" it will see that $expected_v is an SC and so will pass control to the
       Test::Deep::Regexp class by do something like "$expected_v->descend($got_v)". The
       "descend()" method should just return true or false.

       This gives you enough to write your own SCs but I haven't documented how diagnostics works
       because it's about to get an overhaul (theoretically).


       By default, Test::Deep will export everything in its "v0" tag, as if you had written:

         use Test::Deep ':v0';

       Those things are:

         all any array array_each arrayelementsonly arraylength arraylengthonly bag
         blessed bool cmp_bag cmp_deeply cmp_methods cmp_set code eq_deeply hash
         hash_each hashkeys hashkeysonly ignore Isa isa listmethods methods noclass
         none noneof num obj_isa re reftype regexpmatches regexponly regexpref
         regexprefonly scalarrefonly scalref set shallow str subbagof subhashof
         subsetof superbagof superhashof supersetof useclass

       A slightly better set of exports is the "v1" set.  It's all the same things, with the
       exception of "Isa" and "blessed".  If you want to import "everything", you probably want
       to "use Test::Deep ':V1';".

       There's another magic export group:  ":preload".  If that is specified, all of the
       Test::Deep plugins will be loaded immediately instead of lazily.




         Ricardo Signes <>


       Fergal Daly <>, with thanks to Michael G Schwern for Test::More's
       is_deeply function which inspired this.

       Please do not bother Fergal Daly with bug reports.  Send them to the maintainer (above) or
       submit them at the issue tracker <>.


       Copyright 2003, 2004 by Fergal Daly <>.

       This program is free software; you can redistribute it and/or modify it under the same
       terms as Perl itself.